Investigations for arthroplasty device infections

International consensus criteriaParvizi, 2018 and evidence-based algorithms McNally, 2021Muhlhofer, 2017Suren, 2021 support a multimodal approach to the diagnosis of arthroplasty device infections, utilising clinical observations, serum inflammatory markers, and peri-prosthetic sampling, ideally obtained during surgical debridement¹. This approach is particularly useful in late chronic infections as symptoms may be subtle and overlap with those of mechanical failure, and pathogenic organisms are sparsely distributed and overlap with common culture contaminants.

In patients with symptoms or signs of sepsis or septic shock, empirical antibiotic therapy should not be delayed while awaiting surgical peri-prosthetic sampling. Two sets of blood for culture should be taken before antibiotic administration. A joint aspirate should be collected as soon as possible, but should not delay antibiotic administration.

In all other patients having surgery, empirical antibiotic therapy is usually withheld preoperatively to optimise the microbiological yield of surgical peri-prosthetic sampling. For patients with acute symptoms or fever, but who are not suspected to have sepsis or septic shock, also take 2 sets of blood for culture and a joint aspirate. If bacteraemia is confirmed, directed antibiotic therapy should not be delayed while awaiting surgical peri-prosthetic sampling.

Patients with late chronic or indolent infection may have been inappropriately started on antibiotics before surgery. If possible in the investigation of late chronic or indolent infection, surgical peri-prosthetic sampling should occur after an antibiotic-free period.

Surgical antibiotic prophylaxis decreases the risk of surgical site infection and should be administered at the usual time before surgical incision. Surgical antibiotic prophylaxis should not be delayed until after peri-prosthetic samples have been collected. In contrast to preoperative antibiotic therapy, surgical antibiotic prophylaxis given before intraoperative sampling does not significantly reduce the microbiological yieldHansen, 2014Perez-Prieto, 2016Reyes, 2018Tetreault, 2014.

Peri-prosthetic samples collected intraoperatively should include 4 to 6 joint tissue biopsiesDudareva, 2018Osmon, 2013Peel, 2017Zmistowski, 2014 and a sample of synovial fluid. Microscopy, culture and histological examination of these samples are required to detect infection with optimal sensitivity and specificity. The specific number of joint biopsies required depends on multiple factors, including the laboratory methods for processing samples. It is critical that the multidisciplinary team work closely with their microbiology laboratory and understand their procedures. The use of bead milling, incubation of tissue samples in blood culture bottles, and extending incubation to 10 days is likely to increase culture yield. It is important to provide relevant clinical information (eg presence of prosthesis, immune compromise, trauma) on the request form, to assist the laboratory in determining the appropriate tests required.

Sonication of the removed prosthesis can be used to extract organisms from a biofilm. With optimal tissue sampling and processing, routine sonication of the removed prosthesis is unlikely to significantly increase diagnostic sensitivity. Sonication may be of use in selected cases, particularly if antibiotic treatment has been given before surgeryDudareva, 2018Zmistowski, 2014.

Diagnostically challenging cases should be discussed with a clinical microbiologist, in particular when infection is clinically apparent, but there is no growth from initial cultures or if atypical organisms such as mycobacteria, Nocardia or fungi are suspected. These pathogens may require specific culture media or conditions, prolonged incubation periods or molecular methods.